Evaluation of Road Pricing by Travel Behavior Model with Fuzzy Logic

研究的重点是设置道路收费对出行行为的影响。通过序贯估计过程，建立了出行者对交通政策和日常出行模式产生反应的模糊出行行为模型。该模型对于估计具有人类模糊性的出行模式具有一定的优势。因此，从出行模式分析中可以认识到不同类型的影响。采用序贯估计过程设计了模糊出行行为模型来描述道路收费设置中的出行模式。特别是，出行者的选择被确定为总结对交通政策的反应。由于出行者的感知可以用模糊逻辑来表示，因此通过评估过程可以很容易地分析单个出行者的决策。在模型标定过程中，也体现了模糊逻辑公式的另一个优点。出行者对道路收费的反应可以用统计的方法进行总结。以中心区通勤时段的道路收费为例进行了分析。模式转换被认为是道路收费的最大影响。同样，出行路线的变化对城市网络的影响也是第二大的。总体效果的大小可以用与拥堵通行费的价值成比例地选择备选方案的个体数量来衡量。出行模式的变化可以联系道路收费的选项进行讨论。由于出行模式是通过在日常活动之间累积出行来生成的，因此采用了带序贯估计的出行行为描述模型。道路收费的影响可以被描述为道路收费设置前对现有出行的间接影响，表现为出行模式的个体变化。

The study focuses on the change of travel behaviors corresponding to the installation of road pricing. The fuzzy travel behavior model is proposed with the sequential estimation process for the reaction of trip maker to the transport policy and the daily travel pattern production. The model may have the advantages to estimate the trip pattern with human fuzziness. Therefore, the different types of impact would be realized from the travel pattern analysis. The fuzzy travel behavior model is designed with the sequential estimation process to describe the travel pattern in road pricing installation. In particular, the options of trip maker are determined to summarize the reactions to transport policy. Since the perception of trip maker can be formulated with fuzzy logic, the decision of individual trip maker might be easily analyzed through the estimation process. Another advantage of fuzzy logic formulation is realized as well in the process of model calibration. The reactions of trip maker to road pricing can be summarized statistically. The road pricing for commuting time period on the central area is assumed as an example case. Modal shift is regarded as the largest effect of road pricing. Similarly, the change of travel route is observed as the second largest effect on urban network. The scale of overall effect can be measured as the number of individuals with selecting alternative options in proportion to the value of congestion toll. The changes in travel pattern can be discussed connecting to the options for road pricing. As the travel pattern is generated with accumulating trips between daily activities, the travel behavior description model with sequential estimation is applied. The impact of road pricing can be described as individual change of travel pattern to show the indirect influence to the existing trips before road pricing installations.